AI Innovation Marathon: From Bold Ideas to Real-World Impact

Mark Passtoors
Salesforce Marketing Cloud Director, Valtech

July 14, 2025

When we kicked off the Innovation Marathon, our goal was clear: help our clients explore the potential of AI in a way that’s fast, focused and grounded in practical business challenges. We set it up in small teams with clear goals. We planned for eight weeks to build something valuable together. 

Instead of running another round of presentations or one-off pilots, we asked: “What if we created a format where clients and experts could explore bold AI ideas together within a defined, intense sprint?” 

The goal wasn’t to move faster for the sake of speed. It was to go further by focusing, collaborating and making progress together. What followed was one of the most energizing and rewarding programs I’ve ever been part of.  

From curiosity to real-world impact

Each participating client brought a unique challenge to the table. Some focused on internal efficiency. Others looked to improve decision-making or enhance the customer experience. In the first few weeks, we moved rapidly from ideas to working prototypes. 

What made it possible? Short cycles, fast feedback and a shared willingness to build, learn and adapt. This was AI innovation in motion, driven by clarity and intent.  

Here’s a what’s already taking shape:  

  • For a leading cash transit provider, we’re reimagining how they listen to and act on customer feedback, bringing together structured and unstructured data to better understand client needs and respond in real time.  
  • In agriculture, we’re using AI to reduce claims by predicting the quality of strawberry batches early on. By turning environmental signals into quality forecasts, we can avoid mismatches between expected and actual quality. This helps to better match batches to the right buyers, plan inspections more efficiently and prevent lower-quality produce from reaching clients with higher expectations.  
  • For a global dairy trading company, we’ve created a virtual team of AI-powered analysts to monitor and summarize market intelligence, providing traders with a digestible, always-updated knowledge base. 
  • For a leading automotive brand, we’re helping them better understand who’s most likely to book a test drive (with it being the key indicator for conversion), enabling more timely, relevant engagement across the customer journey.
  • For a hospitality and event organization, we’ve built a fully automated translation workflow for their marketing campaigns. With a high volume of multilingual content, translation used to be time-consuming. Now, using generative AI, we preserve their tone of voice, maintain full campaign context and consistency, and drastically reduce time to market from weeks to days. The process is seamless and fully integrated into their workflow. 

These use cases combine Valtech’s internal accelerators, deep industry knowledge, close collaboration with clients, and the capabilities of platformpartners. 

 

A new kind of collaboration 

From day one, we set the tone: this wasn’t just about inspiration. It was about action. 

We kicked off with in-person workshops where client teams met with Valtech experts to define a clear business challenge and identify AI use cases worth exploring. The use cases ranged from customer experience to operational efficiency, from marketing automation to smart prioritization. 

Over 42 days, each team worked toward a working proof of concept. Milestone check-ins gave space for feedback, iteration and shared learning with peers from other participating organizations. 

The structure worked because it was simple, focused and collaborative. Everyone knew the destination: to show the value of AI in a real-world business context. 

The grand finale  

On June 26, we brought everyone back together for the grand finale in Amsterdam. And while it marked the official end of the Innovation Marathon, it felt much more like the beginning of something bigger.  

Each team presented their proof of concept in a fast-paced, 10-minute pitch format. The ideas covered everything from automating content creation to surfacing market insights instantly and even predicting product quality before harvest. These weren’t just concepts. They were working prototypes, designed and built around real use cases.  

What stood out most was where these ideas came from. Not from a top-down directive. Not from a vendor pitch, but from within the teams themselves. People who saw opportunities in their day-to-day and used this program to turn ambition into momentum.  

We were also joined by Professor Paul Iske, who delivered an inspiring keynote on the power of intelligent failure. His message was clear: innovation is rarely linear. Experimentation, iteration and even getting it wrong are essential parts of progress. It was a timely reminder that real innovation comes from trying, learning and trying again.  

 

What we learned  

Every Innovation Marathon team taught us something new. A few key lessons stood out: 

  • The best innovation comes from within. When teams feel safe to explore and are given the space to lead, they’ll uncover more valuable ideas than any outside expert ever could.
  • Start small, deliver fast. The most effective solutions solved tightly scoped problems, and those small wins are now opening doors to bigger conversations. 
  • Progress matters more than polish. Sharing early versions, blockers and imperfect prototypes built more trust and momentum than waiting for the "perfect" reveal. 

What’s next?

The Innovation Marathon wasn’t built as a one-off. It’s part of a growing set of collaboration formats we’re exploring with our clients. Formats that bring business value and experimentation together, quickly. 

We’re already thinking about the next edition. 

So, if you’re sitting on an AI idea, or wondering how to bring it to life, let’s talk. This format works. And the next run might be even better.  

 

Request your seat now